The proposed research in the Hartl laboratory has been to develop E coli and related species for use as experimental organisms in population genetics and evolution. Theoretical models for the distribution and abundance of transposable insertion sequences (IS) in natural isolate of E coli have been developed and tested against extensive data. The evolution of insertion sequences will be studied at the level of DNA sequences within and among host species. The level of variation among three families of IS elements in E coli isolates has already been determined. The proposed continuation consists of four principal projects. (1) To determine whether there may be an exclusion mechanism that prevents the two major subfamilies of IS1 elements (IS1F and IS1R) from occurring together in the same strain. This test involves a novel application of the oligonucleotide ligation assay. (2) To examine the distribution of insertion sites of IS elements in the chromosome of a sample of natural isolates of E coli. The IS elements will be mapped using inverse PCR to obtain the flanking sequences, followed by hybridization to a canonical set of phage clones that cover the entire genome. Evidence for nonrandomness in the distribution of insertion sites in the laboratory strain E coli K-12 has been obtained. We propose the hypothesis that transposition of IS elements tends to occur relatively locally, resulting in clustering of the insertion sites of elements in any particular IS family, but independence among the insertion sites of elements in different families. This hypothesis will be tested. (3) To identify functionally important regions of IS elements that are conserved in evolution, and to assess whether horizontal transfer of IS elements occurs as readily among bacterial species as it does among strains of a single species. These issues will be resolved by determining the nucleotide sequences of IS1, IS3 and IS30 in species of enteric bacterial related to E coli, including copies from Escherichia fergusonii, Escherichia hermanii, Escherichia vulneris, Klebsiella pneumoniae, Shigella dysenteriae, and Serratia odorifera. (4) To determine the amount and type of genetic variation in coding sequences present in the conjugational plasmid F among natural isolates of E coli. Genetic variation is widespread among chromosomal genes in E coli, but the level of sequence variation within each family of IS elements is very low. Genetic variation among plasmid genes is completely unknown. It is our hypothesis that conjugational plasmids should be thought of as accessory bacterial chromosomes, and that the level and distribution of genetic variation among plasmid genes will be found to be very similar to chromosomal genes. In addition, several IS elements have been mapped to the F plasmid present in laboratory strains, and the proposed study will test the hypothesis that different F plasmids often carry different complements of IS elements, which would imply that any particular IS family is not dependent upon a particular type of plasmid for its dissemination among strains.